Impact of reservoir geochemistry on low salinity waterflooding: Global sensitivity analysis

Abstract

Low salinity waterflooding (LSW) has been receiving an increasing attention in the industry as it has several advantages over alternative strategies to enhanced oil recovery (EOR), i.e. i) outperforms conventional EOR methods in terms of additional oil recovery, ii) involves lower chemical costs, iii) it is environmentally friendly, and iv) requires a relatively simple field process implementation. We quantify low salinity effect (LSE) on oil recovery upon employing a mechanistic geochemical model under parametric uncertainty. Our ultimate goal is to characterize the impact of the reservoir rock and fluid properties on the performance of LSW. To this end, we propose a geochemical model of LSW built within a commercial compositional reservoir simulator. The model is designed to account for geochemical processes recently identified through experimental and pore-scale analyses, i.e. double layer expansion, multiple ion exchange and mineral dissolution/precipitation. The geochemical model outputs were first compared with two low salinity coreflood experiments reported in the literature and then applied to predict oil recovery under parametric uncertainty. Our study encompasses the effect of formation water ionic concentrations, rock mineralogy and reservoir temperature on LSW performance. Global sensitivity indices were calculated considering reasonable intervals of variability of the studied parameters. Our analysis focuses on the total oil recovery at three different time scales: 6 months (before water breakthrough), 2 years (after water breakthrough), and 10 years (the end of the injection). The results are discussed in the frame of quantitative screening criteria development, which is necessary to assess whether a given reservoir may be a promising candidate for LSW. Our results indicate that reservoir temperature is the parameter which demonstrated the most significant impact on the LSW performance. Ion concentration also displays a significant influence on oil recovery, and we identify different chemical species in carbonate and sandstone environments. On the contrary, mineral composition shows a limited influence on oil recovery. In particular, our results show that clay presence might be not essential for LSW to be effective in sandstone reservoirs. We discuss the impact of our results in the context of experimental design aimed at an improved constraining of LSW parameterization. Our study suggests that reservoir fluid composition (e.g., concentration of Ca2+, SO42− and Na+), in addition to reservoir temperature, should be prioritized in future experimental campaigns to better understand its influence on LSW under different reservoir and operation conditions.